Hugging Face's logo

ozgraslan
/
diffusion_dino_pointmaze_test

Safetensors
Model card Files
xet
Community
ozgraslan commited on
Commit
abd26ee
·
verified ·
1 Parent(s): a9b6f3a

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +92 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +32 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +64 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +262 -0
config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion_dino",
3
+ "n_obs_steps": 3,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 224,
10
+ 224
11
+ ]
12
+ },
13
+ "observation.goal": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 224,
18
+ 224
19
+ ]
20
+ },
21
+ "observation.state": {
22
+ "type": "STATE",
23
+ "shape": [
24
+ 4
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 2
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "push_to_hub": true,
39
+ "repo_id": "ozgraslan/diffusion_dino_pointmaze",
40
+ "private": null,
41
+ "tags": null,
42
+ "license": null,
43
+ "pretrained_path": null,
44
+ "horizon": 32,
45
+ "n_action_steps": 25,
46
+ "use_goal_conditioning": true,
47
+ "normalization_mapping": {
48
+ "VISUAL": "MEAN_STD",
49
+ "STATE": "MIN_MAX",
50
+ "ACTION": "MIN_MAX"
51
+ },
52
+ "drop_n_last_frames": 5,
53
+ "vision_backbone": "facebookresearch/dinov2",
54
+ "crop_shape": [
55
+ 224,
56
+ 224
57
+ ],
58
+ "crop_is_random": true,
59
+ "pretrained_backbone_weights": "dinov2_vits14",
60
+ "output_key": "x_norm_patchtokens",
61
+ "use_group_norm": false,
62
+ "spatial_softmax_num_keypoints": 32,
63
+ "use_separate_rgb_encoder_per_camera": false,
64
+ "down_dims": [
65
+ 512,
66
+ 1024,
67
+ 2048
68
+ ],
69
+ "kernel_size": 5,
70
+ "n_groups": 8,
71
+ "diffusion_step_embed_dim": 128,
72
+ "use_film_scale_modulation": true,
73
+ "noise_scheduler_type": "DDPM",
74
+ "num_train_timesteps": 100,
75
+ "beta_schedule": "squaredcos_cap_v2",
76
+ "beta_start": 0.0001,
77
+ "beta_end": 0.02,
78
+ "prediction_type": "epsilon",
79
+ "clip_sample": true,
80
+ "clip_sample_range": 1.0,
81
+ "num_inference_steps": null,
82
+ "do_mask_loss_for_padding": false,
83
+ "optimizer_lr": 0.0001,
84
+ "optimizer_betas": [
85
+ 0.95,
86
+ 0.999
87
+ ],
88
+ "optimizer_eps": 1e-08,
89
+ "optimizer_weight_decay": 1e-06,
90
+ "scheduler_name": "cosine",
91
+ "scheduler_warmup_steps": 500
92
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:255902e06c0e6093ab1e6ed69dfeb1c724a0f11fca7035e527fdb959abf34a5a
3
+ size 1124659400
policy_postprocessor.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_postprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "unnormalizer_processor",
6
+ "config": {
7
+ "eps": 1e-08,
8
+ "features": {
9
+ "action": {
10
+ "type": "ACTION",
11
+ "shape": [
12
+ 2
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "MEAN_STD",
18
+ "STATE": "MIN_MAX",
19
+ "ACTION": "MIN_MAX"
20
+ }
21
+ },
22
+ "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
23
+ },
24
+ {
25
+ "registry_name": "device_processor",
26
+ "config": {
27
+ "device": "cpu",
28
+ "float_dtype": null
29
+ }
30
+ }
31
+ ]
32
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:345aac6afe35c5090dd8f517984159989d079a149aec2ea5d10d7bd6327f4ca9
3
+ size 7168
policy_preprocessor.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {}
8
+ }
9
+ },
10
+ {
11
+ "registry_name": "to_batch_processor",
12
+ "config": {}
13
+ },
14
+ {
15
+ "registry_name": "device_processor",
16
+ "config": {
17
+ "device": "cuda",
18
+ "float_dtype": null
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "normalizer_processor",
23
+ "config": {
24
+ "eps": 1e-08,
25
+ "features": {
26
+ "observation.image": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 224,
31
+ 224
32
+ ]
33
+ },
34
+ "observation.goal": {
35
+ "type": "VISUAL",
36
+ "shape": [
37
+ 3,
38
+ 224,
39
+ 224
40
+ ]
41
+ },
42
+ "observation.state": {
43
+ "type": "STATE",
44
+ "shape": [
45
+ 4
46
+ ]
47
+ },
48
+ "action": {
49
+ "type": "ACTION",
50
+ "shape": [
51
+ 2
52
+ ]
53
+ }
54
+ },
55
+ "norm_map": {
56
+ "VISUAL": "MEAN_STD",
57
+ "STATE": "MIN_MAX",
58
+ "ACTION": "MIN_MAX"
59
+ }
60
+ },
61
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
62
+ }
63
+ ]
64
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:345aac6afe35c5090dd8f517984159989d079a149aec2ea5d10d7bd6327f4ca9
3
+ size 7168
train_config.json ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "ozgraslan/pointmaze_umaze_v3_goal_224",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": false,
8
+ "max_num_transforms": 3,
9
+ "random_order": false,
10
+ "tfs": {
11
+ "brightness": {
12
+ "weight": 1.0,
13
+ "type": "ColorJitter",
14
+ "kwargs": {
15
+ "brightness": [
16
+ 0.8,
17
+ 1.2
18
+ ]
19
+ }
20
+ },
21
+ "contrast": {
22
+ "weight": 1.0,
23
+ "type": "ColorJitter",
24
+ "kwargs": {
25
+ "contrast": [
26
+ 0.8,
27
+ 1.2
28
+ ]
29
+ }
30
+ },
31
+ "saturation": {
32
+ "weight": 1.0,
33
+ "type": "ColorJitter",
34
+ "kwargs": {
35
+ "saturation": [
36
+ 0.5,
37
+ 1.5
38
+ ]
39
+ }
40
+ },
41
+ "hue": {
42
+ "weight": 1.0,
43
+ "type": "ColorJitter",
44
+ "kwargs": {
45
+ "hue": [
46
+ -0.05,
47
+ 0.05
48
+ ]
49
+ }
50
+ },
51
+ "sharpness": {
52
+ "weight": 1.0,
53
+ "type": "SharpnessJitter",
54
+ "kwargs": {
55
+ "sharpness": [
56
+ 0.5,
57
+ 1.5
58
+ ]
59
+ }
60
+ },
61
+ "affine": {
62
+ "weight": 1.0,
63
+ "type": "RandomAffine",
64
+ "kwargs": {
65
+ "degrees": [
66
+ -5.0,
67
+ 5.0
68
+ ],
69
+ "translate": [
70
+ 0.05,
71
+ 0.05
72
+ ]
73
+ }
74
+ }
75
+ }
76
+ },
77
+ "revision": null,
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "torchcodec",
80
+ "streaming": false
81
+ },
82
+ "env": {
83
+ "type": "pointmaze_umaze",
84
+ "task": "PointMaze_UMaze_Visual-v3",
85
+ "fps": 10,
86
+ "features": {
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 2
91
+ ]
92
+ },
93
+ "proprio": {
94
+ "type": "STATE",
95
+ "shape": [
96
+ 4
97
+ ]
98
+ },
99
+ "visual": {
100
+ "type": "VISUAL",
101
+ "shape": [
102
+ 224,
103
+ 224,
104
+ 3
105
+ ]
106
+ },
107
+ "target": {
108
+ "type": "VISUAL",
109
+ "shape": [
110
+ 224,
111
+ 224,
112
+ 3
113
+ ]
114
+ }
115
+ },
116
+ "features_map": {
117
+ "action": "action",
118
+ "proprio": "observation.state",
119
+ "visual": "observation.image",
120
+ "target": "observation.goal"
121
+ },
122
+ "max_parallel_tasks": 1,
123
+ "disable_env_checker": true,
124
+ "episode_length": 300
125
+ },
126
+ "policy": {
127
+ "type": "diffusion_dino",
128
+ "n_obs_steps": 3,
129
+ "input_features": {
130
+ "observation.image": {
131
+ "type": "VISUAL",
132
+ "shape": [
133
+ 3,
134
+ 224,
135
+ 224
136
+ ]
137
+ },
138
+ "observation.goal": {
139
+ "type": "VISUAL",
140
+ "shape": [
141
+ 3,
142
+ 224,
143
+ 224
144
+ ]
145
+ },
146
+ "observation.state": {
147
+ "type": "STATE",
148
+ "shape": [
149
+ 4
150
+ ]
151
+ }
152
+ },
153
+ "output_features": {
154
+ "action": {
155
+ "type": "ACTION",
156
+ "shape": [
157
+ 2
158
+ ]
159
+ }
160
+ },
161
+ "device": "cuda",
162
+ "use_amp": false,
163
+ "push_to_hub": true,
164
+ "repo_id": "ozgraslan/diffusion_dino_pointmaze",
165
+ "private": null,
166
+ "tags": null,
167
+ "license": null,
168
+ "pretrained_path": null,
169
+ "horizon": 32,
170
+ "n_action_steps": 25,
171
+ "use_goal_conditioning": true,
172
+ "normalization_mapping": {
173
+ "VISUAL": "MEAN_STD",
174
+ "STATE": "MIN_MAX",
175
+ "ACTION": "MIN_MAX"
176
+ },
177
+ "drop_n_last_frames": 5,
178
+ "vision_backbone": "facebookresearch/dinov2",
179
+ "crop_shape": [
180
+ 224,
181
+ 224
182
+ ],
183
+ "crop_is_random": true,
184
+ "pretrained_backbone_weights": "dinov2_vits14",
185
+ "output_key": "x_norm_patchtokens",
186
+ "use_group_norm": false,
187
+ "spatial_softmax_num_keypoints": 32,
188
+ "use_separate_rgb_encoder_per_camera": false,
189
+ "down_dims": [
190
+ 512,
191
+ 1024,
192
+ 2048
193
+ ],
194
+ "kernel_size": 5,
195
+ "n_groups": 8,
196
+ "diffusion_step_embed_dim": 128,
197
+ "use_film_scale_modulation": true,
198
+ "noise_scheduler_type": "DDPM",
199
+ "num_train_timesteps": 100,
200
+ "beta_schedule": "squaredcos_cap_v2",
201
+ "beta_start": 0.0001,
202
+ "beta_end": 0.02,
203
+ "prediction_type": "epsilon",
204
+ "clip_sample": true,
205
+ "clip_sample_range": 1.0,
206
+ "num_inference_steps": null,
207
+ "do_mask_loss_for_padding": false,
208
+ "optimizer_lr": 0.0001,
209
+ "optimizer_betas": [
210
+ 0.95,
211
+ 0.999
212
+ ],
213
+ "optimizer_eps": 1e-08,
214
+ "optimizer_weight_decay": 1e-06,
215
+ "scheduler_name": "cosine",
216
+ "scheduler_warmup_steps": 500
217
+ },
218
+ "output_dir": "/network/scratch/o/ozgur.aslan/outputs/train/diffusion_dino_pointmaze",
219
+ "job_name": "pointmaze_umaze_diffusion_dino",
220
+ "resume": false,
221
+ "seed": 100000,
222
+ "num_workers": 4,
223
+ "batch_size": 64,
224
+ "steps": 200000,
225
+ "eval_freq": 5000,
226
+ "log_freq": 200,
227
+ "save_checkpoint": true,
228
+ "save_freq": 5000,
229
+ "use_policy_training_preset": true,
230
+ "optimizer": {
231
+ "type": "adam",
232
+ "lr": 0.0001,
233
+ "weight_decay": 1e-06,
234
+ "grad_clip_norm": 10.0,
235
+ "betas": [
236
+ 0.95,
237
+ 0.999
238
+ ],
239
+ "eps": 1e-08
240
+ },
241
+ "scheduler": {
242
+ "type": "diffuser",
243
+ "num_warmup_steps": 500,
244
+ "name": "cosine"
245
+ },
246
+ "eval": {
247
+ "n_episodes": 50,
248
+ "batch_size": 50,
249
+ "use_async_envs": false
250
+ },
251
+ "wandb": {
252
+ "enable": true,
253
+ "disable_artifact": true,
254
+ "project": "lerobot-diffusion-pointmaze",
255
+ "entity": "jakd9",
256
+ "notes": null,
257
+ "run_id": "vtqok7xq",
258
+ "mode": null
259
+ },
260
+ "checkpoint_path": null,
261
+ "rename_map": {}
262
+ }